There is increasing recognition that natural disturbance is a dominant force in forest development. This paper sets out to review natural disturbances and their effects in forests in Australia, and to determine whether or not the effects of management of forests for all of their benefits can be contained within the known effects of natural disturbance. The history and evolutionary significance of fire in Australia is reviewed, and the differing fire ecologies of two representative species, Eucalyptus regnans in the south-east (which is killed by fire and regenerates prolifically from seed after fire) and Eucalyptus marginata in the south-west (which survives all but the most severe fires and regenerates from both shoots and seeds) are outlined. The development of E. regnans following stand-replacing fire can be defined as highly resilient (returning quickly to the pre-disturbance state) and that of E. marginata as resistant (difficult to move from the pre-disturbance state). The topic of forest management in relation to disturbance is discussed within the context of forests of mountain ash (Eucalyptus regnans), the world's tallest flowering plant, in south-eastern Australia. These forests in their natural state are essentially even-aged or at the most include two to three age-classes, the result of regeneration following stand-replacing bushfires. They produce timber of high value, and cuffent land planning and management prescriptions allow for about one-half of the forest area to be managed for timber production on a sustained-yield basis. The silvicultural system is clear-felling, followed by burning to give stand-replacing conditions. Mountain ash regenerating from this system is one of the fastest growing tree species in the world and at its best has a volume increment of 50 m3 ha-1 year-1 and net primary production of 35 t ha-1 year-1 over the first 10 years. There is no definitive evidence of a loss of diversity of plant species in timber production forests. The greatest problem is to manage land for timber production and for the diversity of fauna which depend on nesting hollows in large dead or living trees. However, all of the species found in older forests can be found in regrowth forests which have been burned or logged over the past 50 years and which include large dead or living trees with nesting hollows. Them has been speculation that harvesting mountain ash forests by clearfelling and regenerating them following slash-burning will lead to a loss of nutrients and a consequent loss of productivity. Nutrient cycling in mountain ash is shown to be resilient to disturbance. Rapid uptake of nutrients by the regenerating forest, immobilization of nutrients by microorganisms, and increased rates of nitrogen fixation are processes which lead to the conservation of nutrients following stand-replacing fires. There is no evidence of productivity decline following bushfire or timber harvesting. Another concern about forest harvesting is that there will be a reduction in carbon storage. The break-even point for E. regnans plantations yielding short-lived products (e.g. paper) is 3 7 years and for E. regnans forests grown for sawn timber, 60 years. Harvesting forests on rotations of 80-120 years will therefore result in an increase in carbon storage; however, it would take several rotations to restore carbon storage equivalent to that of old-age forest. Much of the literature on disturbance illustrates the difficulties of dealing with traditional and sometimes deeply entrenched concepts of how plant communities develop and the nature of the end-state of development. Development of mountain ash forest fits within Egler's model of initial floristic composition; there is no stable and self-perpetuating end-point, the climax is probably no more than the state where species change is immeasurably slow, and maximum diversity and productivity are maintained by random periodic disturbance. Does this set of conclusions provide the basis for management of the forest? The management of diversity in heath is discussed as an example; the issue is less contentious than in forests since no commercial product is harvested. Coastal heath on soils of poor fertility in Victoria also follows the model of initial floristic composition; diversity and productivity are maintained by fire, decreasing markedly within 20-50 years of disturbance. Management can use fire to maintain a mosaic of regenerated heathland to accommodate the range of responses of fauna and flora to disturbance, thereby maximizing diversity. The solution of maintaining diversity through an intermediate level of disturbance is well-established in the ecological literature; through disturbance, an equilibrium is never reached and higher diversity is maintained. In the managed forest, management dictates the disturbance regime (frequency, size and intensity) which must be fined to the attributes, or life histories, of the organisms to be managed. This paper presents the view that timber harvesting in Australian forests is ecologically sustainable and that the effects of management can be contained within the framework of those caused by natural disturbance. Given that each plot in the forest differs from all other plots and that none is at steady state, a solution for the management of diversity is to use the whole of the forest estate (parks, stream reserves, catchment reserves, old-age forest, forests of different ages resulting from past fires and logging) so that diversity of the estate, rather than diversity of each plot, is maximized.
